The Fundamental Theorem of Algebra: a Visual Approach
Total Page:16
File Type:pdf, Size:1020Kb
The Fundamental Theorem of Algebra: A Visual Approach Daniel J. Velleman Department of Mathematics and Statistics Amherst College Amherst, MA 01002 In the real numbers, a linear equation ax + b = 0 (with a 6= 0) always has a solution. Quadratic equations don't always have real solutions, but the quadratic formula gives solutions in the complex numbers. What if we move on to polynomial equations of higher degree? Might we need to extend the complex numbers to some larger number system in order to find solutions to these equations? The answer is no: solutions to all polynomial equations can be found in the complex numbers. This is the content of the Fundamental Theorem of Algebra: Fundamental Theorem of Algebra. Every nonconstant polynomial with complex coefficients has a root in the complex numbers. Some version of the statement of the Fundamental Theorem of Algebra first appeared early in the 17th century in the writings of several mathematicians, including Peter Roth, Albert Girard, and Ren´eDescartes. The first proof of the Fundamental Theorem was published by Jean Le Rond d'Alembert in 1746 [2], but his proof was not very rigorous. Carl Friedrich Gauss is often credited with producing the first correct proof in his doctoral dissertation of 1799 [15], although this proof also had gaps. (For a comparison of these two proofs, see [26, pp. 195{200].) Today there are many known proofs of the Fundamental Theorem of Algebra, including proofs using methods of algebra, analysis, and topology. (The references include many papers and books containing proofs of the Fundamental Theorem; [14] alone contains 11 proofs.) Our focus in this paper will be on the use of pictures to see why the theorem is true. Of course, if we want to use pictures to display the behavior of polynomials defined on the com- plex numbers, we are immediately faced with a difficulty: the complex numbers are two-dimensional, so it appears that a graph of a complex-valued function on the complex numbers will require four dimensions. Our solution to this problem will be to use color to represent some dimensions. We begin by assigning a color to every number in the complex plane. Figure 1 is a picture of the complex plane in which every point has been assigned a different color. The origin is colored black. Traveling counterclockwise around a circle centered at the origin, we go through the colors of a standard color wheel: red, yellow, green, cyan, blue, magenta, and back to red. Points near the origin have dark colors, with the color assigned to a complex number z approaching black as z approaches 0. Points far from the origin are light, with the color of z approaching white as jzj approaches infinity. Every complex number has a different color in this picture, so a complex number can be uniquely specified by giving its color. 1 30 20 10 0 -10 -20 -30 -30 -20 -10 0 10 20 30 Figure 1: Assigning a color to each point in the complex plane. 2 We can now use this color scheme to draw a picture of a function f : C ! C as follows: we simply color each point z in the complex plane with the color corresponding to the value of f(z). From such a picture, we can read off the value of f(z), for any complex number z, by determining the color of the point z in the picture, and then consulting Figure 1 to see what complex number is represented by that color. For example, Figure 2 is a picture of the function f(z) = z3. Three things are immediately evident in this picture. First, we see that the center of the picture is very dark. This is because when z is small, z3 is very small, and therefore the color assigned to z3 is very dark. Second, the colors fade out quickly when we move toward the outside of the picture. This is because when z is large, z3 is very large, and therefore its color is very light. But what is most striking about the picture is that when we go counterclockwise around a circle centered at the origin, we go through the colors of the color wheel three times. This illustrates the fact that the argument of z3 is three times the argument of z, and therefore the image of a circle centered at the origin under the cubing function wraps around the origin three times. As an illustration of how such a picture can help us understand a function, note that it is immediately evident from Figure 2 that every nonzero complex number has three cube roots. For example, the color assigned to the number 1 in Figure 1 is a deep red. Therefore, the three cube roots of 1 are the three points in Figure 2 that are colored this particular shade of red. Let us consider now a more complicated function. Figure 3 is a picture of the polynomial f(z) = z8 − 2z7 + 2z6 − 4z5 + 2z4 − 2z3 − 5z2 + 4z − 4. Perhaps the first thing one notices in this picture is that the Fundamental Theorem of Algebra does, indeed, hold for f. Since the color assigned to the number 0 is black, the roots of f appear in this picture as six black dots. The fact that the polynomial has degree eight also shows up in the picture. For large z, the z8 term in f(z) dominates the other terms, and therefore the outer parts of the picture look similar to a picture of the function z8: the colors begin to fade toward white as we move toward the edges of the picture, but before they fade out we can see that, as we go around the picture counterclockwise, the colors of the color wheel are repeated eight times. Why does f, a polynomial of degree eight, have only six roots? The reason is that two of the roots are double roots, andp this fact is also evident in the picture.p The single roots occur at the points −1, 2, and (−1 ± i 7)=2, and the double roots are at (1 ± i 3)=2. The regions around the double roots are somewhat darker than those around the single roots, and at the double roots the colors of the color wheel wrap around the root twice, whereas at the single roots they wrap around only once. In general, if a polynomial f has a root of multiplicity k at a point z0, then when f(z) is expanded in powers of z − z0 it will have the form k f(z) = c(z − z0) + (higher degree terms): For z close to z0, the first term in this expansion will dominate the higher degree terms, and therefore k we have f(z) ≈ c(z − z0) . Thus, near the point z0, the picture of f will be similar to the picture of the function czk near 0. In particular, the colors of the color wheel will wrap around the point z0 k times, and the larger k is, the darker the picture will be near z0. (It is not hard to see that the effect of the coefficient c on the picture is to alter the darkness of the picture near z0, and also to rotate the arrangement of colors around z0. For example, near the root at −1 in Figure 3, the colors have been rotated 180 degrees, so that red is to the left of the root rather than to the right. 3 6 4 2 0 -2 -4 -6 -6 -4 -2 0 2 4 6 Figure 2: f(z) = z3. 4 2 1 p 0 -1 -2 -1 0 1 2 Figure 3: f(z) = z8 − 2z7 + 2z6 − 4z5 + 2z4 − 2z3 − 5z2 + 4z − 4. 5 This is because when the polynomial f(z) of Figure 3 is expanded in powers of z + 1, the coefficient of z + 1 is a negative real number.) We might describe this situation by saying that at a single root, f is \locally linear," at a double root it is \locally quadratic," etc. In fact, a similar principle applies even at points that are not roots, although this is a little harder to see in our pictures. For any complex number z0, by expanding f(z) in powers of z − z0 we can find complex numbers b and c and a positive integer k such that k f(z) = b + c(z − z0) + (higher degree terms): k It follows that for z near z0 we will have f(z) ≈ b + c(z − z0) . We have already seen that the k picture of the function c(z − z0) has a black dot at z0, with the colors that surround 0 in Figure 1 wrapping around z0 k times. The effect of adding b to this function is to \shift" the colors in the color space of Figure 1 from 0 to b. The result is that z0 will be colored with the color assigned to b, and it is the colors surrounding b in Figure 1 that will wrap around z0 k times. (The fact that all of the colors surrounding b appear near z0 illustrates that the Open Mapping Theorem holds for f.) For example, in Figure 3 we have drawn a small circle at the point representing the complex number −:7 + :8i, and labeled this point p. The point p is colored light green, which is the color representing the complex number f(−:7+:8i) ≈ −7:5+15:8i. Moving up from p, the colors become greenish-yellow and then yellow; moving down, they shift toward cyan.